Originally posted by Tom-
Quote:
I have a database of many factors.
What would you suggest I do to start a regression study?
Look at single factors, combinations?
I could use Excel to plot finsh positions or beaten lengths of horses with certain factors by rank, look for postivie correlation?
Is there a better program out there to run multiple studies?
WhatI do now is to querry Access for things like
Running Style F or E
Qirin speed points >6
F1 velocity rank 1
EP velocity rank <3
Then figure out impact values and roi for the results.
It sounds like you guys are skinng the cat with a sharper knife.
|
Tom,
I've spent hundreds of, no- make that
thousands of- man hours, over the last decade or so, coding out my own programs- always tinkering with my own database and the interface to it. I use Visual Basic and SQL.
One thing that I managed to do very well was create a good set of tools for myself. Take a single factor, say best figure last race. Say I want to know how that factor performs across x number of races or for x time period. Instead of writing and running a single query to test how the top rated horse in that category performed, and then writing and running a second query to see how the second ranked horse in that category performed, and then writing and running a third query to see how the third ranked horse in that category performed, etc- I instead took the time to develop and test a VB function that hits the database and returns the results of positions 1-20 by rank within that single category. So instead of 20 separate trips to the database, I click a couple of drop downs and buttons and make ONE TRIP through the database to get back information on how each of the 20 positions (by rank) performed in that category.
I also have another VB function that does the same thing by numeric ranges and difference from category leader instead of by rank. Again, the output I get is that for 20 separate numeric ranges (as opposed to one single range) for a single databse trip or query.
Having stuff like that (once you've taken the time to create it) is a GREAT timesaver.
Sounds like you are an Access user. If you are so inclined you could accomplish the same thing by creating and storing queries in Access.
Personally, I never liked that approach. Even though SQL Query Analyzer (in my case) packs a strong punch in this area, I opted, instead, to write VB code to hit the database, store the output in public variables, and output those to a textbox when done. The numbers I generate are often of my own making. This approach seemed to give me better ability to tweak my own numbers as the need arose. You could also do the same thing in Access using VBA code if you are familiar with it.